Driver scoring system and method using optimum path deviation

Abstract

Techniques are disclosed to determine driver scoring that consider deviations in map and sensor data in driver score computations. The deviations can be based on deviations from an optimum driving path, and include the determination of vectors between a reference point and one or more landmarks, and one or more reference vectors between the reference point the landmark(s) when traveling on an optimum driving path. A difference between the vectors may then be used determine the deviations from the optimum driving path. In contrast to the conventional approaches, the use of positional deviations (e.g. determined from road markings) in the computation of driver scores allows for improved driver scoring techniques and driver characterizations.

The invention claimed is: 
     
         1 . A controller of a vehicle, comprising:
 a communication interface configured to receive, at a sampling frequency, map data and sensor data, and   processing circuitry configured to, while navigating a driving environment:
 determine a positional deviation between a landmark included in the map data and a landmark included in the sensor data by:
 determining, based on the sensor data, a position vector between a reference point of the vehicle and the landmark included in the sensor data; 
 determining, based on the map data, a reference position vector between the reference point and the landmark included in the map data, the reference position vector being associated with an optimum path of the vehicle; and 
 calculating a vector difference between the position vector and the reference position vector to determine the positional deviation; 
 selectively adjusting the sampling frequency based upon the determined positional deviation; 
 determining, based upon a comparison of the determined positional deviation to a threshold value, whether to include the determined positional deviation as part of a driver score calculation; and 
 calculating a driver score, based on the determined positional deviation and a positive determination to include the determined positional deviation as part of the driver score calculation, to provide a continuous path monitoring to improve driver scoring accuracy, the driver score thereby quantifying a deviation of the vehicle from the optimum path. 
 
   
     
     
         2 . The controller of  claim 1 , wherein the positional deviation is a deviation of the vehicle from an ideal position of an autonomous vehicle determined based on the map data. 
     
     
         3 . The controller of  claim 1 , wherein the sensor data comprises one or more images, videos, or video frames captured by a camera of the vehicle. 
     
     
         4 . The controller of  claim 1 , wherein the processing circuitry is configured to adjust the sampling frequency of the sensor data based on one or more driving events to generate sampled sensor data, the driver score being based on a deviation between the map data and the sampled sensor data. 
     
     
         5 . The controller of  claim 1 , wherein the processing circuitry is further configured to determine a driving event based on the sensor data, the driver score being determined based on the determined positional deviation and the driving event. 
     
     
         6 . The controller of  claim 5 , wherein the driving event comprises an acceleration event, a velocity event, a braking event, a directional-change event, a distracted-driver event, impaired-driver event, and/or a collision-avoidance event. 
     
     
         7 . The controller of  claim 5 , wherein the sensor data is generated from a sensor of the vehicle and/or a sensor within or attached to the vehicle that is communicatively coupled to the controller. 
     
     
         8 . The controller of  claim 5 , wherein the sensor data includes data from a mobile device within the vehicle. 
     
     
         9 . The controller of  claim 1 , wherein the landmark included in the map data and/or the sensor data comprises: a lane marking, a road edge, a road arrow, a stop line, a cross walk, a traffic sign or light, or utility pole. 
     
     
         10 . The controller of  claim 1 , wherein the driver score is determined based further on map metadata. 
     
     
         11 . The controller of  claim 1 , wherein the processing circuitry is configured to:
 determine a running windowed average of vector differences for two or more landmarks over a path of the vehicle; and   determine the driver score based on the running windowed average of vector differences.   
     
     
         12 . The controller of  claim 1 , wherein the processing circuitry is configured to dynamically adjust a number of landmarks used to determine the positional deviation based on one or more characteristics of a driving segment. 
     
     
         13 . The controller of  claim 1 , wherein the processing circuitry is configured to dynamically adjust a number of position vectors and/or reference position vectors used to determine the vector difference based on one or more characteristics of a driving segment. 
     
     
         14 . The controller of  claim 1 , wherein: the reference point is a center of view of an image included in the sensor data, and
 a position of the reference point changes with respect to the landmark as the vehicle moves.   
     
     
         15 . A controller of a vehicle, comprising:
 a communication interface configured to receive sensor data at a sampling frequency; and   processing circuitry configured to, while navigating a driving environment:
 determine a landmark associated with a drivable path based on the sensor data; 
 determine, based on the sensor data, a position vector between a current position of the vehicle on the drivable path and the determined landmark; 
 determine, based on map data, a reference position vector between the current position and the determined landmark, the reference position vector being associated with an optimum path of the vehicle; 
 determine a vector difference between the position vector of the determined landmark and the reference position vector of the determined landmark; 
 selectively adjust the sampling frequency based upon the determined vector difference; 
 determine, based upon a comparison of the determined vector difference to a threshold value, whether to include the determined vector difference as part of a driver score calculation; and 
 calculate a driver score, based on the vector difference and a positive determination to include the determined vector difference as part of the driver score calculation, to thereby provide a continuous path monitoring to improve driver scoring accuracy, the driver score quantifying a deviation of the vehicle from the optimum path. 
   
     
     
         16 . The controller of  claim 15 , wherein the processing circuitry is configured to compare the determined vector difference with the threshold value, the driver score being determined based on the comparison. 
     
     
         17 . The controller of  claim 15 , wherein the communication interface is further configured to receive the reference position vector of the landmark. 
     
     
         18 . The controller of  claim 15 , wherein the processing circuitry is configured to determine the reference position vector based on map data. 
     
     
         19 . The controller of  claim 18 , wherein the communication interface is further configured to receive the map data. 
     
     
         20 . A controller of a vehicle, comprising:
 interface means for receiving sensor data at a sampling frequency; and   processing means for, while navigating a driving environment:
 determining a landmark associated with a drivable path based on the sensor data; 
 determining, based on the sensor data, a position vector between a current position of the vehicle on the drivable path and the determined landmark; 
 determining, based on map data, a reference position vector between the current position and the determined landmark, the reference position vector being associated with an optimum path of the vehicle; 
 calculating a vector difference between the position vector of the determined landmark and the reference position vector of the determined landmark; 
 selectively adjusting the sampling frequency based upon the calculated vector difference; 
 determining, based upon a comparison of the calculated vector difference to a threshold value, whether to include the calculated vector difference as part of a driver score calculation; and 
 calculating a driver score, based on the vector difference and a positive determination to include the calculated vector difference as part of the driver score calculation, to thereby provide a continuous path monitoring to improve driver scoring accuracy, the driver score quantifying a deviation of the vehicle from the optimum path.